R Markdown

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.

Load the libraries

library(tidyverse)
## ── Attaching packages ───────────────────────────────────────────────────────────────────────────────────── tidyverse 1.2.1 ──
## ✔ ggplot2 3.1.1       ✔ purrr   0.3.2  
## ✔ tibble  2.1.1       ✔ dplyr   0.8.0.1
## ✔ tidyr   0.8.3       ✔ stringr 1.4.0  
## ✔ readr   1.3.1       ✔ forcats 0.4.0
## Warning: package 'ggplot2' was built under R version 3.5.2
## Warning: package 'tibble' was built under R version 3.5.2
## Warning: package 'tidyr' was built under R version 3.5.2
## Warning: package 'purrr' was built under R version 3.5.2
## Warning: package 'dplyr' was built under R version 3.5.2
## Warning: package 'stringr' was built under R version 3.5.2
## Warning: package 'forcats' was built under R version 3.5.2
## ── Conflicts ──────────────────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
library(RCurl)
## Warning: package 'RCurl' was built under R version 3.5.2
## Loading required package: bitops
## 
## Attaching package: 'RCurl'
## The following object is masked from 'package:tidyr':
## 
##     complete
library(dplyr)
library(readr)
library(scales)
## 
## Attaching package: 'scales'
## The following object is masked from 'package:purrr':
## 
##     discard
## The following object is masked from 'package:readr':
## 
##     col_factor
library(ggthemes)
## Warning: package 'ggthemes' was built under R version 3.5.2

Load the data

test3 <- getURL("https://raw.githubusercontent.com/fivethirtyeight/data/master/marriage/divorce.csv")
divorcedata <- read.csv(text = test3)

First attempt

This was an attempt at finding a code to make the y scale a percent. The code tested was (This code is not in a code chunk because the file will not knit wiht in incorrect code): scale_y_continous(labels=scales::percent)

This code chunk was on trying to find a way to convert data to percents

help(percent)
HS_percent <- percent(divorcedata$HS_3544)
divorcedata$HS_percent = divorcedata$HS_3544*(100)

The follwing are attempts at converting data to percents. They did not list how they come up with “completed college” variable, so i had to try various different ways and then graph them to see if i got the same line that they did.

divorcedata$HS_percent = divorcedata$HS_3544*(100)
divorcedata$HSP = divorcedata$HS_3544*(100) 
divorcedata$CSTotal = divorcedata$BAp_3544 + divorcedata$BAo_3544 + divorcedata$GD_3544
divorcedata$CSTotal2 = divorcedata$BAp_3544 + divorcedata$BAo_3544
divorcedata$CSP2 = divorcedata$CSTotal2*(100)
divorcedata$CSP = divorcedata$CSTotal*(100)
divorcedata$SCP = divorcedata$SC_3544*(100)

I am attempting at plotting just one line with a percent. Failed due to scale_y_ continous: HS <- ggplot(divorcedata) + geom_line(mapping = aes(x = year, y = HS_3544)) HS + scale_y_continous(labels=scales::percent)

This is an attempt at plotting one line

HS <- ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HS_3544))

Keeping code that i wanted to try to use to plot the percent, and the limit i wanted to set it too

scale_y_continuous(name = "percent", limits = c(0, 25))
## <ScaleContinuousPosition>
##  Range:  
##  Limits:    0 --   25

Here I tested out the scaling code with the plotting code. It was not working.

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HS_3544)) + scale_y_continuous(name = "percent", limits = c(0, 25))

This is when i tried to make the graph of divorce rates by education

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HS_3544))

I had to do further testing of various percents to match theirs. I wish more infomration had been given on how they went from teh dataset to the variables in the graph.

divorcedata$HS_percent = divorcedata$HS_3544*(100)
divorcedata$HSP = divorcedata$HS_3544*(100) 
divorcedata$CSTotal = divorcedata$BAp_3544 + divorcedata$BAo_3544 + divorcedata$GD_3544
divorcedata$CSTotal2 = divorcedata$BAp_3544 + divorcedata$BAo_3544
divorcedata$CSP2 = divorcedata$CSTotal2*(100)
divorcedata$CSP = divorcedata$CSTotal*(100)
divorcedata$SCP = divorcedata$SC_3544*(100)

Succesful maping with one percent. now time to add in more lines

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HSP)) +
  scale_y_continuous(name = "percent", limits = c(0, 25))

Figuring out how to get colors to match the best

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HSP, color = "HSP")) +
  geom_line(mapping = aes(x = year, y = SCP, color = "Some College")) +
  geom_line(mapping = aes(x = year, y = CSP2, color = "all college")) +
  scale_y_continuous(name = "percent", limits = c(0, 25))

The final percents for the other two variables

divorcedata$HSP = divorcedata$HS_3544*(100) 
divorcedata$SCP = divorcedata$SC_3544*(100)
divorcedata$BAP = divorcedata$BAp_3544*(100)

Trying to get labels

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HSP), color = 'blue') +
  geom_line(mapping = aes(x = year, y = SCP), color = 'light blue') +
  geom_line(mapping = aes(x = year, y = BAP), color = 'purple') +
  scale_y_continuous(name = "percent", limits = c(0, 25)) +
  ggtitle("Divorce Rates by Education: Ages 35-44") +
  scale_color_discrete(name = "education level", breaks=c("HSP", "SCP", "BAP"), labels=c("High School", "Some College", "Bachelors")) +
  theme(legend.position = "right")

More attempts at labeling on graph (not included in a code chunk because file will not knit) ggplot(divorcedata) + geom_line(mapping = aes(x = year, y = HSP), color = ‘blue’) + geom_line(mapping = aes(x = year, y = SCP), color = ‘light blue’) + geom_line(mapping = aes(x = year, y = BAP), color = ‘purple’) + scale_y_continuous(name = “percent”, limits = c(0, 25)) + ggtitle(“Divorce Rates by Education: Ages 35-44”) + scale_color_discrete(name = “education level”) + geom_dl(aes(label = Education), method = list(dl.combine(“first.points”, “last.points”), cex = 0.8)

still working, but can’t get the labels exaclty how I would like them

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HSP, label = "High School"), color = 'blue') +
  geom_line(mapping = aes(x = year, y = SCP), color = 'light blue') +
  geom_line(mapping = aes(x = year, y = BAP), color = 'purple') +
  scale_y_continuous(name = "percent", limits = c(0, 25)) +
  ggtitle("Divorce Rates by Education: Ages 35-44") 
## Warning: Ignoring unknown aesthetics: label

FINAL PRODUCT BELOW

AND FINALLY ALL THE PIECES TOGETHER. The code below produced my most successful figure! I also attempted to match the colors to the one online, which only used shades of blue.

ggplot(divorcedata) +
  geom_line(mapping = aes(x = year, y = HSP), color = 'cyan') +
  geom_line(mapping = aes(x = year, y = SCP), color = 'navy blue') +
  geom_line(mapping = aes(x = year, y = BAP), color = 'light blue') +
  scale_y_continuous(name = "percent", limits = c(0, 25)) +
  labs(title = "Divorce Rates by Education", subtitle = "Ages 35-44", x = "years") +
  annotate("text", x= 1990, y = 23, label = "Some college") +
  annotate("text", x= 2000, y = 13, label = "High school or less") +
  annotate("text", x= 2005, y = 5, label = "College Graduates") +
  theme_fivethirtyeight() +
  geom_segment(aes(2000, 6, xend = 2000, yend =10.1)) +
  geom_segment(aes(2000, 15, xend = 2000, yend = 17)) +
  geom_segment(aes(1990, 17, xend = 1990, yend = 21))

The original plots and article can be found at: https://fivethirtyeight.com/features/marriage-isnt-dead-yet/